pflogd — packet filter logging daemon
pflogd [−DragonFly] [−d delay] [−f filename] [−s snaplen] [expression]
pflogd is a background daemon which reads packets logged by pf(4) to the packet logging interface pflog0 and writes the packets to a logfile (normally /var/log/pflog) in tcpdump(8) binary format. These logs can be reviewed later using the −r option of tcpdump(8), hopefully offline in case there are bugs in the packet parsing code of tcpdump(8).
pflogd closes and then re-opens the log file when it receives SIGHUP, permitting newsyslog(8) to rotate logfiles automatically. SIGALRM causes pflogd to flush the current logfile buffers to the disk, thus making the most recent logs available. The buffers are also flushed every delay seconds.
If the log file contains data after a restart or a SIGHUP, new logs are appended to the existing file. If the existing log file was created with a different snaplen, pflogd temporarily uses the old snaplen to keep the log file consistent.
pflogd tries to preserve the integrity of the log file against I/O errors. Furthermore, integrity of an existing log file is verified before appending. If there is an invalid log file or an I/O error, logging is suspended until a SIGHUP or a SIGALRM is received.
The options are as follows:
Debugging mode. pflogd does not disassociate from the controlling terminal.
Time in seconds to delay between automatic flushes of the file. This may be specified with a value between 5 and 3600 seconds. If not specified, the default is 60 seconds.
Log output filename. Default is /var/log/pflog.
Analyze at most the first snaplen bytes of data from each packet rather than the default of 96. The default of 96 is adequate for IP, ICMP, TCP, and UDP headers but may truncate protocol information for other protocols. Other file parsers may desire a higher snaplen.
Check the integrity of an existing log file, and return.
Selects which packets will be dumped, using the regular language of tcpdump(8).
Process ID of the currently running pflogd.
Default log file.
Log specific tcp packets to a different log file with a large snaplen (useful with a log-all rule to dump complete sessions):
# pflogd -s 1600 -f suspicious.log port 80 and host evilhost
Display binary logs:
# tcpdump -n -e -ttt -r /var/log/pflog
Display the logs in real time (this does not interfere with the operation of pflogd):
# tcpdump -n -e -ttt -i pflog0
Tcpdump has been extended to be able to filter on the pfloghdr structure defined in 〈 net/if_pflog.h〉 . Tcpdump can restrict the output to packets logged on a specified interface, a rule number, a reason, a direction, an IP family or an action.
Address family equals IPv4.
Address family equals IPv6.
Interface name equals "kue0".
Interface name equals "kue0".
Rule number equals 10.
Reason equals match. Also accepts "bad-offset", "fragment", "short", "normalize" and "memory".
Action equals pass. Also accepts "block".
The direction was inbound.
The direction was outbound.
Display the logs in real time of inbound packets that were blocked on the wi0 interface:
# tcpdump -n -e -ttt -i pflog0 inbound and action block and on wi0
The pflogd command appeared in OpenBSD 3.0.
Can Erkin Acar
BSD July 9, 2001 BSD